Security and Privacy Issues and Solutions for UAVs in B5G Networks – A Review

Unmanned Aerial Vehicles (UAVs) are increasingly integrated with Beyond Fifth Generation (B5G) networks to support intelligent applications such as surveillance, disaster management, smart transportation, and military operations. Despite their advantages, UAV-enabled B5G networks face significant security and privacy challenges due to their high mobility, open wireless medium, and resource-constrained architecture. Common threats include flooding, blackhole, sybil, and wormhole attacks, which can severely degrade network performance and compromise sensitive data.

This paper presents a comprehensive review of security and privacy issues in UAV-assisted B5G networks, emphasizing the role of machine learning-based intrusion detection systems (IDS). Network flow-level features such as packet rate, throughput, byte rate, and packet drop rate are analyzed to detect anomalous UAV communication behavior. Various classification models, including K-Nearest Neighbors, Multi-Layer Perceptron, Random Forest, LightGBM, and XGBoost, are reviewed and compared based on detection accuracy and robustness. Experimental observations indicate that ensemble-based models, particularly XGBoost and Random Forest, provide high detection accuracy and reliable performance for real-time UAV traffic classification.

Furthermore, this study highlights practical mitigation strategies such as secure routing, node authentication, and anomaly isolation to enhance network resilience. The review concludes that intelligent IDS frameworks integrated with B5G infrastructure are essential for achieving secure, privacy-preserving, and reliable UAV communications in future wireless networks.

Keywords

UAV Security, B5G Networks, Intrusion Detection System, Machine Learning, Network Traffic Analysis, Privacy Preservation